SortNet: Learning to Rank by a Neural Preference Function

Rigutini, Leonardo; Papini, Tiziano; Maggini, Marco; Scarselli, Franco

doi:10.1109/tnn.2011.2160875

Cited by 52 publications

(31 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, several works exist [4,5,23,30] that have proposed neural network architectures for learning-to-rank. We do not focus on a specific network architecture in this paper, but instead propose a new training criterion for learning-to-rank from implicit feedback that in principle allows unbiased network training for a large class of architectures.…”

Section: Related Workmentioning

confidence: 99%

A General Framework for Counterfactual Learning-to-Rank

Agarwal

Takatsu

Zaitsev

et al. 2019

Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval

102

117

View full text Add to dashboard Cite

Implicit feedback (e.g., click, dwell time) is an attractive source of training data for Learning-to-Rank, but its naive use leads to learning results that are distorted by presentation bias. For the special case of optimizing average rank for linear ranking functions, however, the recently developed SVM-PropRank method has shown that counterfactual inference techniques can be used to provably overcome the distorting effect of presentation bias. Going beyond this special case, this paper provides a general and theoretically rigorous framework for counterfactual learning-to-rank that enables unbiased training for a broad class of additive ranking metrics (e.g., Discounted Cumulative Gain (DCG)) as well as a broad class of models (e.g., deep networks). Specifically, we derive a relaxation for propensity-weighted rank-based metrics which is subdifferentiable and thus suitable for gradient-based optimization. We demonstrate the effectiveness of this general approach by instantiating two new learning methods. One is a new type of unbiased SVM that optimizes DCG -called SVM PropDCG -, and we show how the resulting optimization problem can be solved via the Convex Concave Procedure (CCP). The other is Deep PropDCG, where the ranking function can be an arbitrary deep network. In addition to the theoretical support, we empirically find that SVM PropDCG significantly outperforms existing linear rankers in terms of DCG. Moreover, the ability to train non-linear ranking functions via Deep PropDCG further improves performance.

show abstract

Section: Related Workmentioning

confidence: 99%

A General Framework for Counterfactual Learning-to-Rank

Agarwal

Takatsu

Zaitsev

et al. 2019

Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval

102

117

View full text Add to dashboard Cite

show abstract

“…RankNet adjusts the attribute weights to best meet pairwise user choices [134] where the implicit feedback such as clickthrough and other user interactions is treated as vector of features which is later integrated directly into the ranking algorithm. SortNet is a pairwise learning method [135] with its associated priority function provided by a multi-layered neural network with a feed-forward configuration and is is trained in the learning phase with a dataset formed of pairs of documents where the associated score of the preference function is provided. SortNet is based on the minimization of the square error function between the network outputs and preferred targets on every unique couple of documents.…”

Section: Neural Network In Learning To Rank Algorithmsmentioning

confidence: 99%

Neural Networks in Big Data and Web Search

Serrano

2018

Data

View full text Add to dashboard Cite

As digitalization is gradually transforming reality into Big Data, Web search engines and recommender systems are fundamental user experience interfaces to make the generated Big Data within the Web as visible or invisible information to Web users. In addition to the challenge of crawling and indexing information within the enormous size and scale of the Internet, e-commerce customers and general Web users should not stay confident that the products suggested or results displayed are either complete or relevant to their search aspirations due to the commercial background of the search service. The economic priority of Web-related businesses requires a higher rank on Web snippets or product suggestions in order to receive additional customers. On the other hand, web search engine and recommender system revenue is obtained from advertisements and pay-per-click. The essential user experience is the self-assurance that the results provided are relevant and exhaustive. This survey paper presents a review of neural networks in Big Data and web search that covers web search engines, ranking algorithms, citation analysis and recommender systems. The use of artificial intelligence (AI) based on neural networks and deep learning in learning relevance and ranking is also analyzed, including its utilization in Big Data analysis and semantic applications. Finally, the random neural network is presented with its practical applications to reasoning approaches for knowledge extraction.

show abstract

“…A similar approach called SortNet is proposed by Rigutini et al (2011). They introduce a three-layered network architecture called CmpNN, which takes two input object vectors and outputs two real-valued numbers.…”

Section: Comparison With Existing Nn-based Approachesmentioning

confidence: 99%

Dyad ranking using Plackett–Luce models based on joint feature representations

Schäfer

Hüllermeier

2018

Mach Learn

View full text Add to dashboard Cite

Label ranking is a specific type of preference learning problem, namely the problem of learning a model that maps instances to rankings over a finite set of predefined alternatives. Like in conventional classification, these alternatives are identified by their name or label while not being characterized in terms of any properties or features that could be potentially useful for learning. In this paper, we consider a generalization of the label ranking problem that we call dyad ranking. In dyad ranking, not only the instances but also the alternatives are represented in terms of attributes. For learning in the setting of dyad ranking, we propose an extension of an existing label ranking method based on the Plackett-Luce model, a statistical model for rank data. This model is combined with a suitable feature representation of dyads. Concretely, we propose a method based on a bilinear extension, where the representation is given in terms of a Kronecker product, as well as a method based on neural networks, which allows for learning a (highly nonlinear) joint feature representation. The usefulness of the additional information provided by the feature description of alternatives is shown in several experimental studies. Finally, we propose a method for the visualization of dyad rankings, which is based on the technique of multidimensional unfolding.

show abstract

SortNet: Learning to Rank by a Neural Preference Function

Cited by 52 publications

References 24 publications

A General Framework for Counterfactual Learning-to-Rank

A General Framework for Counterfactual Learning-to-Rank

Neural Networks in Big Data and Web Search

Dyad ranking using Plackett–Luce models based on joint feature representations

Contact Info

Product

Resources

About